Game theoretic pricing and optimal routing in optical networks

In this paper, we investigate a game theory based model for pricing, wavelength and flow assignment in multiwavelength optical networks. The scenario includes several backbone long-haul optical network transport service providers (NTSP), offering their services in terms of bandwidth to Internet service providers (ISPs). We present a microeconomic flow control technique that models the NTSP as non-cooperative competitors for the ISP bandwidth demands and the routing and wavelength allocation as a multicommodity flow optimization problem. We model the competition using Bertrand duopoly model where the competitors compete by lowering prices until they reach their marginal costs as price offered for unit bandwidth. The simulations show that though the non-cooperative market environment the networks can achieve profits and allocate the bandwidth requests. We provide a comparison in terms of network profit compared to the profitability of the same networks but using fixed pricing.